K8s Monitoring with Prometheus + cAdvisor DaemonSet

1. Run cAdvisor Daemonset which K8s deploys the cAdvisor on every node in cluster

kubectl create -f https://raw.githubusercontent.com/veerendra2/prometheus-k8s-monitoring/master/cadvisor-daemonset.yml

2. Run Prometheus Configmap which provides config for prometheus server. If nessasary edit the config and run

kubectl create -f https://raw.githubusercontent.com/veerendra2/prometheus-k8s-monitoring/master/prometheus-configmap.yml

3. Apply RBAC

kubectl apply -f https://raw.githubusercontent.com/veerendra2/prometheus-k8s-monitoring/master/prometheus-rbac.yml

4. Now create Deployment which K8s creates Prometheus POD, Service with a NodePort 30900

kubectl create -f https://raw.githubusercontent.com/veerendra2/prometheus-k8s-monitoring/master/prometheus-deployment.yml

• Open TCP port 30900 on any node in the cluster.
• Access the Prometheus UI http://<NODE IP>:30900

NOTE

• I have relabeled __meta_kubernetes_pod_name to container_label_io_kubernetes_pod_name and __meta_kubernetes_namespace to container_label_io_kubernetes_pod_namespace which look like cAdvisor's metric tag. So carefull it will add exported_ for those two metrics like exported_container_label_io_kubernetes_pod_namespace and exported_container_label_io_kubernetes_pod_name

• run.sh - https://raw.githubusercontent.com/veerendra2/prometheus-k8s-monitoring/master/run.sh

• kill.sh - https://raw.githubusercontent.com/veerendra2/prometheus-k8s-monitoring/master/kill.sh